This is the second and final submission of the competitive renewal of R01 MH077139, A Large-Scale Schizophrenia Association Study in Sweden. This project has been very successful with all aims accomplished and, indeed, exceeded. We successfully ascertained the numbers of cases and controls for which we were funded (N~12,996 in total). Instead of candidate gene genotyping, we conducted GWAS on around half the sample. These data were key components in two International Schizophrenia Consortium Nature papers and are a sizable fraction of the Psychiatric GWAS Consortium schizophrenia mega-analysis. Finally, our Swedish sample is being used for two grand opportunity NIH grants (exome sequencing and genome-wide methylomics). We propose here new aims that use this unique sample to increase our knowledge of SCZ and to keep this highly productive team intact. The samples studied here will increase available samples for SCZ by 61% which will provide substantially enhanced power for detection of networks and loci that confer risk or protection. In Aim 1, we will complete the genomic characterization of the full sample of 12,996 subjects (5,851 SCZ cases, 7,145 controls). This includes GWAS genotyping and a novel Illumina exome chip on all subjects plus exome sequencing on 1,000 cases and 1,000 controls. All assays are funded by philanthropy. We request funds for our team to conduct rigorous quality control, imputation, and data warehousing. In Aim 2, we will conduct analyses designed to elucidate genetic effects that confer risk or protection for SCZ. (2a) Pathway and network analyses: identify additional high-confidence pathways for SCZ by rigorous assessment of whether genomic results are enriched for smaller p-values in genes in classical and novel empirical pathways. (2b) Conduct the first highly powered genome-scale analysis of exome polymorphism in SCZ. The novel exome chip was developed by Illumina and our colleagues at the Broad, and captures missense and nonsense coding variation with MAF e0.1%, and a substantial fraction with MAF 0.01-0.1%. This could identify low-frequency missense or nonsense polymorphisms that contribute to SCZ. (2c) Analyze directly genotyped and imputed common variants in the Swedish data, and meta-analyze with the PGC SCZ results. Conduct verification genotyping as required. Anticipated outcome: genome-wide significant evidence for 10-20 novel loci. (2d) Identify structural variation contributing to SCZ. Conduct the first systematic evaluations of CNPs, indels, novel insertions, and other structural polymorphisms. Evaluate CNV associations with SCZ, and meta-analyze with external results. Conduct verification genotyping as required. Anticipated outcome: increase understanding of the major CNV loci, and the role of novel structural variants. In Aim 3, estimate if high-confidence genetic effects are modified by established epidemiological risk factors for SCZ. For this complex trait, environmental risks may generally interact with causal genetic variation. For Aim 2 results, are genetic effects strengthened or lessened if epidemiological risk factors are modeled? Risk factors are from high-quality databases and were often assessed prospectively (season of birth, urban/rural birth, paternal and grandparental age, pre-morbid cognitive attainment, and head trauma). Anticipated outcome: understanding of how specific genetic and environmental risk factors interact in SCZ that could lead to new mechanistic hypotheses. Our goal is rapidly to learn more about the genetics of schizophrenia and how genes and environment might act and interact to alter disease risk. We propose to do this using the largest and most comprehensively studied sample collection in the field and which includes both genetic and environmental risk factors.